Copilot Memory now follows JetBrains work across sessions. Check what survives
News
MAI-Code-1.1-Flash costs a quarter request in Copilot. The old model leaves September 10
Google put sign-language AI into Gboard. The server boundary still matters
Nemotron 3.5 Lightning is a 66 GB model with 3B active. Do not confuse compute with memory
Spotify’s AI Persona badge labels the artist, not the song
Claude watermark arrives before the detector. Treat it as a signal, not proof
DeepSeek V4 Flash’s cheapest price is a cache hit. Budget the miss
Claude raised a Riemann-zeta lower bound to 67.2%. The proof package is the story
Meta Muse Glimmer fits in 17 GB. Day-one backend support is the real test
An AI agent rewrites its safety harness after failure. The held-out test matters